Squamous cell carcinoma of the head and neck (SCCHN) represents4.5% of the incident cancers in the USA each year and is the sixth most common cancer worldwide. Despite being a potentiallycurable malignancy in its early stages, the majority of patientspresent with locally advanced disease (stages III-IV) and will die within 2 years of diagnosis aftertreatment with standard approaches. Platinum-based therapy has historically been considered an important chemotherapeutic regimen for SCCHN, particularly in combination with radiotherapy. The nucleotide excision repair (NER) pathway has been implicated as an important factor in modulating overall cancer survival and responsiveness to platinum based chemotherapy agents in a variety of tumor types. We have developed a SNP selection strategy for screening DNA repair pathway haplotypes based on genotyping 384 SNPs using a custom designed Illumina GoldenGate assay. Our SNP selection strategy incorporates haplotype tagSNPs, functional SNPs characterized by amino acid substitutions, evolutionary conservation, thermodynamic tolerance, published epidemiological data and a minimum minor allele frequency (MAF) of 10%. This study was conducted in 539 cases with SCCHN. The mean age of SCCHN diagnosis among these patients was 59.6 years. The overall follow-up time in months (mean±SD, range) was 26.9+23.2 (0.06-95.9) and the progression-free survival was 15.6+17.3 months (1.0-86.3). The most common (~30%) chemotherapy regimen included platinum-based therapy either alone or in combination. Haplotypes were reconstructed for each gene and association with overall survival was evaluated using the Cox proportional hazards model. The most frequent haplotype was used as the referent group for the estimation of hazard ratios (HR). Hazard ratios were calculated for common haplotypes i.e. those that were present at a frequency greater than 2%. The hazard ratios for the haplotypes were adjusted for age, gender, smoking and stage. There were two NER gene haplotypes associated with overall survival in the cases irrespective of treatment modality. The ERCC2 gene haplotype conferred an increased HR of 3.38 (1.17-9.78) whereas the ERCC5 gene conferred a protective effect with an HR of 0.47 (0.28-0.80). In cases treated with platinum based chemotherapy ERCC3 and PARP1 haplotypes were each associated with a 3 fold increase in HR and therefore decreased overall survival. These results support the hypothesis that DNA repair pathways represent important cellular processes relevant to variation in disease survival and treatment response. These results should be interpreted cautiously since adjustments for multiple testing have not yet been applied in this interim data analysis and our results are currently being validated in a larger study population.

Citation Information: In: Proc Am Assoc Cancer Res; 2009 Apr 18-22; Denver, CO. Philadelphia (PA): AACR; 2009. Abstract nr 4843.

100th AACR Annual Meeting-- Apr 18-22, 2009; Denver, CO